Abstract WP120: Portable Bedside Low-Field Magnetic Resonance Imaging Acute Infarct Detection on Floor Level Acute Ischemic Stroke Patients

Abstract

Introduction: FDA approval of Hyperfine has led to interest in using low-field Magnetic Resonance Imaging (MRI) in acute ischemic stroke (AIS). Portable MRI (pMRI) has practical advantages over high-field MRI (hMRI), and prior studies have evaluated pMRI predominantly in the ICU settings. We sought to determine performance of pMRI on floor level AIS patients. Methods: We retrospectively reviewed floor status AIS patients who had pMRI obtained during their hospitalization. We reviewed pMRI for ability to visualize confirmed infarcts on hMRI. hMRI was considered gold standard. Incomplete or technically limited scans were excluded. Our institutional pMRI (Hyperfine) is 0.064 Tesla and our protocol sequences include diffusion-weighted imaging (DWI), Apparent diffusion coefficient, and fluid-attenuated inversion recovery (scan time 24 minutes). Infarct size was measured based upon maximum longitudinal axis on MRI DWI. Results: Among 12 AIS patients, the mean age was 58.5 years (range, 25 to 88) and 7 (58.3%) were female. Mean time from hMRI to pMRI was 46.2 hours (range, -0.9 to 126.9 hrs). Mean Time from LKN to pMRI was 3.6 days (range, 0.87 to 7.8 days). In most 9/12 (75%) patients, pMRI demonstrated acute infarcts (7/12 pMRI scans demonstrated all infarcts seen on hMRI, while 2/12 pMRI scans identified some but not all infarcts). Three 3 (25%) pMRI DWI were negative. On 12 hMRI scans, 15 infarct lesions were demonstrated. Infarct distribution was as follows: cortical (9), deep (1), medullary (1), thalamus (3), and cerebellar (1). Among the 15 infarcts demonstrated on hMRI, 10/15 (67%) were also visible on pMRI. The mean size of infarcts that were not visible on pMRI was 7.7 mm, compared to pMRI visible infarct mean size of 21.7mm (p 0.0436). The smallest infarct that was visible on the pMRI was 7.7mm. Conclusions: pMRIs performed reasonably well on floor-level AIS patients and was able to detect the majority of acute infarcts seen on hMRI. There was a correlation between infarct size and visibility on pMRI, with pMRI performing better on larger sized infarcts. Understanding limitation of pMRI will allow clinical teams to maximize patient selection for use in AIS population.